Security audit

DoctorClaw Weekly Report

Security checks for vulnerabilities and agentic risk

Overview

This skill openly helps generate weekly productivity reports from user-selected task, calendar, and email sources, with disclosed delivery and archive options.

Before installing, decide exactly which task source, calendars, email account, delivery channel, recipients, and archive location it may use. Avoid enabling recurring reports or team/client delivery until those choices are explicit, since the report can combine sensitive work and communication details in one place.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill instructs the agent to scan emails and calendars, compile highlights, and send/archive reports, but it does not warn users that potentially sensitive personal, client, or business data will be processed and stored. In this context, the missing disclosure is dangerous because the generated report can aggregate confidential information into a single artifact and then transmit it to external channels or save it persistently, increasing exposure if permissions, recipients, or storage locations are wrong.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The invocation examples are broad enough that an agent may trigger this skill on generic requests like status checks or scheduling without clearly confirming scope, data sources, or recipients. Because the skill accesses tasks, email, and calendar data and may also schedule recurring execution, underspecified activation increases the risk of unintended collection, summarization, and sharing of sensitive information.

Static analysis

No suspicious patterns detected.